o-Carbonyl-assisted alkaline hydrolyses of methyl benzoates - Journal

Publication Date: July 1968. ACS Legacy Archive. Cite this:J. Am. Chem. Soc. 1968, 90, 16, 4410-4413. Note: In lieu of an abstract, this is the articl...
0 downloads 0 Views 535KB Size
4410

o-Carbonyl-Assisted Alkaline Hydrolyses of Methyl BenzoateslV2 Melvin S. Newman and Arie L. Leegwater Contribution from the Evans Chemistry Laboratory, The Ohio State University, Columbus, Ohio 43210. Received January 15, 1968 Abstract: The rates of alkaline hydrolysis of methyl 6-methyl-2-benzoylbenzoate,methyl 6-methyl-2-acetylmethyl 6-chloro-2-acetylbenzoate,and the 6-unsubstituted analogs benzoate, methyl 6-chloro-2-benzoylbenzoate, have been measured. The rates are reported in Table 11. In each case the 6-substituted esters hydrolyze at rates greater than do the corresponding unsubstituted esters. The results are explained by assuming a rate-controlling intramolecular attack of the anion, formed by adding hydroxyl ion to the ketonic carbonyl group, on the ester grouping which has been forced out of the plane of the aromatic ring by the 6-substituent.

I

n earlier work, the greater rate of alkaline hydrolysis of methyl 6-methyl-2-benzoylbenzoate as compared to that of methyl 2-benzoylbenzoate was explained3 by assuming that the hydrolysis of the 6-methyl compound was initiated by attack of a hydroxide ion on the ketonic carbonyl group followed by intramolecular displacement of methoxide ion. The steric acceleration was raiionalized by postulating that the 6-methyl group forces the ester grouping to be noncoplanar with the benzene ring to which it is attached and hence better oriented for intramolecular attack by the adjacent negative oxygen.

&w COR

A 1.2

=

H: R

=

CnH;

VII; 2 = C l ; - R = CaHj VIII, Z = C1; R = CH3 I X , Z = C1; R = H

there is a steric acceleration when Z is methyl and chloro in the cases where R is methyl and phenyl. The inability to prepare pure VI and IX is unusual as treatment of keto acids with diazomethane gives pure normal ester in all other cases in our experience. Large rate increases in hydrolyses of dimethylphosphoacetoin4 and of methyl 2-formylbenzoate5 have been explained by postulating attack of hydroxide ion on adjacent acetyl4 and formyl gr0ups.j Later, other examples of increased rates of hydrolysis of esters due to attack of hydroxide on adjacent carbonyl groups have appeared.6)' The magnitude of acceleration of alkaline hydrolysis of all the compounds mentioned4-' was considerably greater than the rate increase originally observed here. However, two factors should be mentioned in comparing the results: the compounds having large rate increases4-' all involved addition to a formyl or an acetyl group, and in no case was the steric effect of a group adjacent to the ester function studied. Therefore, we decided to determine the rates of hydrolysis of the compounds represented by the general formula A, where Z would be H, CH3, and C1 and R would be C6H6, CH3, and H. Unfortunately, we were unable to prepare pure samples of VI and IX so that our objectives were only attained in part. However, the work done shows that (1) This research was supported in part by Grant DA-ARO-D-31124-G846 of the U. S . Army Research Office, Durham, N. C., and by Grant GP-5552Xl of the National Science Foundation. (2) This research is reported in greater detail in the Ph.D. thesis of A. L. Leegwater presented to The Ohio State University, 1967. (3) M. S. Newman and S. Hishida, J . A m . Chem. Soc., 84, 3582 (1962). (4) F. Ramirez, B. Hansen, and N. B. DeSai, ibid., 84, 4588 (1962). (5) M. L. Bender and M. S . Silver, ibid., 84, 4589 (1962). (6) Y.Shalitin and S . A. Bernhard, ibid., 86, 2292 (1964). (7) C. N. Lieske, E. G. Miller, Jr., J. J. Zeger, and G . M. Steinberg, ibid., 88, 188 (1966).

Journal of the American Chemical Society

90:16

1 July 31,

Experimental Section8 The normal (I) and pseudo methyl esters of 2-benzoylbenzoic acid used were pure preparations on hand.9 Methyl 6-methyl-2-benzoylbenzoate (IV), bp 158-159" (0.8 mm), la and 6-chloro-2-benzoylbenzoic acid11 were prepared as described. The melting point behavior of the latter acid was confirmed.11 Treatment with ethereal diazomethane yielded pure methyl 6-chloro-2-benzoylbenzoate (VII), bp 190-192" (1.2 mm), infrared absorption at 5.75 (1740) and 6.00 p (1670 cm-1). Anal. Calcd for CljHI1C1O3: C. 65.6; H, 4.0. Foundm: C, 65.4; H , 4.0. 6-Chloro-2-acetylbenzoicAcid. A finely ground mixture of 10.0 g of 3-chlorophthalic anhydride" and 6.7 g of dry malonic acid was heated on a steam bath for 2 hr with 7 ml of dry pyridine.'* After adding 50 ml of water and 4 ml of concentrated HC1 the solution deposited 5.0 g of colorless crystals, mp 124126". Recrystallization from benzene yielded 4.1 g (38%) of pure 6-chloro-2-acetylbenzoic acid, mp 128-129", ir bands at 2.90 (3450) and 5.70 ~ ( 1 7 5 0 cm-'). From the original aqueous filtrate 4.7 g of 3-chlorophthalic acid, mp 184-186", was recovered. Anal. Calcd for C,HiC103: C, 54.5; H, 3.6; C1, 17.8. Foundg: C, 54.3; H, 3.6; CI, 17.8. The above structure was confirmed by decarboxylation (Cu powder, quinoline at reflux for 30 min) and oxidation of the result(8) The term "worked up in the usual way" means that an etherbenzene solution of the products was extracted with dilute NaHCOI solution and/or dilute HC1 as needed, washed with saturated salt solution, filtered through anhydrous MgSOd, and concentrated by distillation or on a rotary evaporator. Analyses marked with a superscript g by the Galbraith Microanalytical Laboratories, Knoxville, Tenn., and superscript m by MicroAnalysis, Inc., Wilmington, Del. (9) M. S . Newman and C. I . Courduvelis, J . Org. Chem., 30, 1795 (1965). (10) M . S . Newman and C. D. McCleary, J . A m . Chern. Soc., 63, 1537 (1941). (11) M. S . Newman and P. G. Scheurer, ibid., 78, 5004 (1956). (12) Compare H . L. Yale, ibid., 69, 1547 (1947).

1968

441 1 ing ketone to m-chlorobenzoic acid with sodium hypoiodite. The melting point of 156-158" was not depressed by mixing with an authentic sample. Esterification of 6-chloro-2-acetylbenzoicacid with diazomethane afforded the pure normal ester VI11 (92z), bp 107-109° (0.5 mm), ir bands a t 5.78 (1730) and 5.90 I.( (1695 cm-l), and with methanolicHCI for 3 hr, the pseudo ester (69$), mp 79.5-81.0", ir band a t 5.65 p (1770cm-l). Anal. Calcd for Cl~HIDCIOl:C, 56.5; H, 4.3; C1, 16.7. Founds (normal ester): C, 56.4; H, 4.2; C1,16.4. Found (pseudo ester): C, 56.5; H,4.3; C1, 16.5. Methyl 2-Acetylbenzoate (11). Treatment of pure 2-acetylbenzoic acid13 with ethereal diazomethane yielded the pure ester 11, bp 134135" (10 mm), 94-95" (0.6 mm),14 ir bands at 5.78 (1730) and 5.90 1.1 (1695 cm-l). 6-Methyl-2-acetylhenzoicAcid. A finely ground mixture of 28.0 of 3-methylphthalic anhydride and 20.8 g of dry malonic acid was heated with 17 ml of dry pyridine for 5 hr on a steam bath.12 After a work-up similar t o that described above for 6-chloro-2-acetylbenzoic acid, 10.2 g (33 %) of pure 6-methyl-2-acetylbenzoic acid, mp 126-127", was obtained. In addition 31 of 3-methylphthalic acid was recovered. By the addition of methylmagnesium iodide to 3-methylphthalic anhydride, there was obtained a 31 yield of 6-methyl-2-acetylbenzoic acid. l5 Esterification with diazomethane yielded pure methyl 6-methyl-2-acetylbenzoate (V), bp 107-108° (0.7 mm), ir bands at 5.75 (1740) and 5.90 p (1695 cm-l), and with methanolic HC1 a 62% yield of pure pseudo methyl ester, mp 72.573.5", ir bands at 5.65 p (1770 cm-l), nmr peaks at 7 2.75 (3 H, aromatic), 7.01 (3 H, singlet, OCHI), 7.31 (3 H, singlet, ArCH3), and 8.24 (3 H, singlet, CHa), was0 btained. This compound must be the pseudo ester and is different from the compound, mp 170180", previously reportedl5 as the pseudo ester. Anal. Calcd for CllH1203: C, 68.7; H, 6.3. Foundm (normal ester): C, 68.8; H, 6.4. Found (pseudo ester): C, 68.7; H, 6.4.. Methyl 2-formylbenzoatel~(111), bp 98-100" (1.7 mm), was prepared by treatment of phthalaldehydic acid with diazomethane. Both 6-methyl-2-formylbenzoic and 6-chloro-2-formylbenzoic acids were prepared for the first time. However, we were unable to isolate pure samples of their normal methyl esters as the acids existed mainly in the cyclic form and gave mixtures of normal and pseudo methyl esters on treatment with diazomethane. To our knowledge, these are the first cases in which appreciable amounts of pseudo esters are formed by diazomethane. Treatment of the silver salts with methyl iodide yielded mixtures of normal and pseudo methyl esters which were not readily separated by chromatography. 6-Methyl-2-formylbenzoic Acid. A suspension of 100 g (0.83 mol) of 2,6-dimethylaniline13 in 250 ml of 48% hydrobromic acid was stirred at 0-5" while a solution of 58 g (0.84 mol) of sodium nitrite in 100 ml of water was slowly added. The resulting solution was added during 30 min to a boiling solution of 66 g of cuprous bromide and 67 ml of 48% hydrobromic acid. The product was steam distilled and worked up in the usual way to afford 70.1 g (46%) of colorless 2,6-dimethyIbromoben~ene,~~ bp 53-55" (2.3 mm). Carbonation of the Grignard reagent formed by treatment of 70 g of the bromo compound and 83 g of ethyl bromide in 400 ml of ether with excess magnesium yielded 45 g (79%) of 2,6-dimethylbenzoic acid,'* mp 118-120". The methyl ester (diazomethane) boiled at 68-70" (2 mm). An illuminated (300-W lamp) mixture of 13.4 g of methyl 2,6dimethylbenzoate and 14.6 g of N-bromosuccinimide in 250 ml of CCla was held at reflux for 2 hr. After removing the succinimide by filtration, the solvent was distilled and the remaining oil treated with 20 ml of methanol and 100 ml of 20% sodium hydroxide at reflux for 20 hr. After acidification and the usual work-up there

x

(13) Obtained from the Aldrich Chemical Co. (14) R. Riemschneider, H . G. Kaahn, and L. Horner, Monatsh. Chem., 91,1040 (1960). (15) P. R. Jones and P. J. Desio, J. Org. Chem., 30, 4293 (1965). (16) E. L. Eliel and A. W. Burgstahler, J . Am. Chem. Soc., 71, 2251 (1949), report bp 136-138" (13 mm). (17) G. Vander Stouw, Ph.D. Dissertation, The Ohio State University, 1964, p 45, reported a 3 4 z yield of 2,6-dimethylbromobenzene, bp 64-67" ( 5 mm). (18) B. van Zanten and W. Th. Nauta, Rec. Trau. Chim., 79, 1216 (1960), report mp 116" for the acid, bp 98-100" (12 mm), for the methyl ester, and bp 90-93 (20 mm) for 2,6-dimethylbromobenzene.

was obtained 10.1 g (83%;) of 7-methylphthalide, mp 83-85', ir band a t 5.71 p (1750 crn-'). The nmr spectrum (CC13 was consistent with the assigned structure. Anal. Calcd for CDHe02: C, 72.9; H, 5.4. Foundm: C, 72.6; H, 5.4. Addition during 30 min of 15.0 g of 3-methylphthalic anhydride t o a refluxing suspension of 2.8 g of LiAIH4 in 300 ml of ether followed by 3 hr at reflux afforded 7-methylphthalide in 69% yield. This route does not establish the structure whereas the first route does. An illuminated (300-W lamp) mixture of 5.5 g df 7-methylphthalide and 6.6 g of N-bromosuccinimide in 150 ml of CC14 was held at reflux for 2 hr. After filtration of the succinimide, the solvent was removed on a rotary evaporator. The oily residue was heated for 10 hr with 75 ml of water. Recrystallizatidhs of the crude solid from benzene afforded 2.2 g (36x1 of 6-methyl-2-formylbenzoic acid, mp 112.5-114.0", ir bands at 3.0 (3340) and 5.73 I.( (1745 cm-1). Anal. Calcd for CgH.@3: C, 65.9; H, 4.9. Founds: C, 65.6; H , 4.8. 6-Chloro-2-formylbenzoic Acid. Reduction of 3-chlorophthalic anhydride" with LiAIH4in ether as described above for 3-methylphthalic anhydride afforded 7-~hlorophthaIide,~~ mp 146-147", ir band a t 5.70 p (1755 cm-]), in 44% yield. Bromination, as above described for 7-methyIphthalide, followed by hydrolysis yielded 6-chloro-2-formylbenzoic acid, mp 187-189", ir band a t 5.75 I.( (1740cm-I), in 47% yield. Anal. Calcd for C8HsC103: C, 52.1; H, 2.7; C1, 19.2. Foundm: C, 52.3; H, 2.9; C1, 18.9. Measurement of Rates of Hydrolysis of Esters. The hydroxide ion catalyzed hydrolysis of the methyl esters in 70:30 (v/v) waterdioxane at 30" was carried out with a thermostated aeckman DU spectrophotometer. The reaction occurred in a thermostated buffer solution in a 3 - d , glass-stoppered quartz cell. The rate constants were determined by following the increase in carboxylate anion absorption with time at 310 mp.20 Experiments with I1 showed that Beer's law was obeyed in the concentration ranges used, The p H of the water-dloxanez1 solutions used in this work was determined with a Beckman Model 76 expanded-scale p H meter with a calomel-fiber junction electrode (39170) and a glass electrode (41260, p H 0-14, Type E-2). The p H meter was standardized with aqueous solutions, as recommended. The glass electrode gives the correct p H at the concentration used in the dioxane-water mixtures.23 A stock solution of each buffer was prepared from aqueous buffers (,u = 0.15 t o 0.19) and the appropriate amount of dioxane. The p H of this stock solution was determined and of the reaction solution after a kinetic run. The p H remained constant within experimental error. After the cell compariment had come to thermal equilibrium, the reaction was initiated by adding 0.25 p1 of ester (neat) or of a solution of ester in dioxane to the buffer solution. The optical density of the reaction mixture was measured against a blank of the appropriate buffer solution at convenient intervals until the optical density remained unchanged. The second-order rate expression, ks[OH-][ester], was reduced to the expression ko[ester]since the hydroxide ion concentrations were held constant by buffers in the 7.93-11.58 pH range. Rewritten in terms of optical density measurements the expression below was used to obtain the rate constant.24

-kot/2.303

=

log (OD, - OD,)

- log OD,

A plot of log (OD, - OD,) us. time (in minutes) gave a straight line for several half-lives. The slope (koi2.303) of this line was determined by the method of least squares. Because we do not know the autoprotolysis constant of water in 30% dioxane, the values for the rate constants listed are not strictly accurate. However, since we were interested mainly in relative (19) S. Biniecki, M . Moll, and L. Rylski, Ann. Pharm. France, 16, 421 (1958), report mp 146-147'; J. Tirouflet, Compt. Rend., 238, 2246 (1954), mp 149". (20) For an ester of each class studied, the rate was shown to be independent of the wavelength employed. (21) Thep-dioxane was purified as described in L. F. Fieser, "Experiments in Organic Chemistry," D . C. Heath, New York, N. Y . , 1941, p 369. (22) R. G. Bates, J . Res. Natl. Bur. Std., 66A, 179 (1962). (23) H . P. Marshall and E. Grunwald, J . Chem. Phys., 21, 2143 (1953). (24) A. A. Frost and R. T. Pearson, "Kinetics and Mechanism," 2nd ed, John Wiley and Sons, Inc., New York, N. Y.,1961, p 38

Newman, Leegwater

i

Alkaline Hydrolyses of Methyl Benzoates

4412

a:;; - &-

rate constants we did not attempt t o determine K, in 3 0 x dioxane and assumed it t o be the same as in pure water. A typical run is shown in Table I. The constants given in Table I1 represent the averages of a t least two runs.

?

+

OD t

5 10 15 20 25 30 34

0.242 0.268 0.293 0.314 0.333 0.349 0.362 0.377 0.389 0.396 0.409 0.415 0.419 0.425 0.431 0.435 0.443

40

46 50 56 60 65 70 75 80 90

OD,

OD mea-

Table 11. Summary of Kinetic Data Relative Rate Constantsa Ester

I I1 I11 IV VI1 V

VI11 I (4-benzoyl) I (pseudo ester)

2.97 4.50 2.63 12.8 17.3 6.33 1.88 4.33 7.28

f 0.20 f 0.12 f 0.02 f 0.07 f 0.00 f 0.17 f 0.03 f 0.27 f 0.30

PH

knK, X 1014 M-1 sec-l

11.58 9.95 7.93 11.58 10.84 9.95 8.70 11.58 11.58

0.078 5.05 310 0.336 2.51 7.10 37.7 0.114 0.192

krel 1 .o 64.8 (1) 3970 4.31 32.2 91 .O (1.4) 483 (7.4) 1.46 2.46

0 At 30.0 f 0.1 O in 70 :30 water-dioxane. These rate constants represent the average of two or three determinations. They were obtained from the constants of the type shown in Table I by dividing by 60. cIf K, in 30% dioxane is the same as in water, the kz values in this column are the true rate constants, kz.

Discussion of Results Before discussing the results of hydrolysis, assumptions must be made concerning the various equilibria and rates involved. Hence the following mechanism is proposed (eq 1 and 2).26 We assume that the observed rate constant is the product of the rate of cyclization of C (kl,eq 2) by the equilibrium constant, K, for the reactions indicated in eq 1. If the formation of the anion C were the ratedetermining step in the hydrolysis we can see no reason why all of the rates for the 6-substituted esters should be greater than those for the corresponding unsubstituted ester. Furthermore, attack of hydroxide ion on the ester function can also be ruled out as the rate-determining step because the rates for ortho-substituted benzoates (in which the substituent cannot participate) would be expected to be slower than the rates for (25) While this work was being finished a similar mechanism was proposed by I